File size: 2,885 Bytes
d766458 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 | # Copyright 2021 DeepMind Technologies Limited
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Functions for building the input features for the AlphaFold model."""
import os
from typing import Mapping, Optional, Sequence
from absl import logging
from colabdesign.af.alphafold.common import residue_constants
from colabdesign.af.alphafold.data import parsers
import numpy as np
# Internal import (7716).
FeatureDict = Mapping[str, np.ndarray]
def make_sequence_features(
sequence: str, description: str, num_res: int) -> FeatureDict:
"""Constructs a feature dict of sequence features."""
features = {}
features['aatype'] = residue_constants.sequence_to_onehot(
sequence=sequence,
mapping=residue_constants.restype_order_with_x,
map_unknown_to_x=True)
features['between_segment_residues'] = np.zeros((num_res,), dtype=np.int32)
features['domain_name'] = np.array([description.encode('utf-8')],
dtype=np.object_)
features['residue_index'] = np.array(range(num_res), dtype=np.int32)
features['seq_length'] = np.array([num_res] * num_res, dtype=np.int32)
features['sequence'] = np.array([sequence.encode('utf-8')], dtype=np.object_)
return features
def make_msa_features(
msas: Sequence[Sequence[str]],
deletion_matrices: Sequence[parsers.DeletionMatrix]) -> FeatureDict:
"""Constructs a feature dict of MSA features."""
if not msas:
raise ValueError('At least one MSA must be provided.')
int_msa = []
deletion_matrix = []
seen_sequences = set()
for msa_index, msa in enumerate(msas):
if not msa:
raise ValueError(f'MSA {msa_index} must contain at least one sequence.')
for sequence_index, sequence in enumerate(msa):
if sequence in seen_sequences:
continue
seen_sequences.add(sequence)
int_msa.append(
[residue_constants.HHBLITS_AA_TO_ID[res] for res in sequence])
deletion_matrix.append(deletion_matrices[msa_index][sequence_index])
num_res = len(msas[0][0])
num_alignments = len(int_msa)
features = {}
features['deletion_matrix_int'] = np.array(deletion_matrix, dtype=np.int32)
features['msa'] = np.array(int_msa, dtype=np.int32)
features['num_alignments'] = np.array(
[num_alignments] * num_res, dtype=np.int32)
return features |